Patterned semiconductor device
By using a single additional lithography stage and selective etching technology in semiconductor devices, combined with the interleaved layer structure, the contradiction between the thick metallized layer on electrical performance and mechanical robustness is solved, and low Rds(on) and mechanical robustness is achieved. Combination reduces passivation rupture and pattern offset, and improves the thermal mechanical stress stability of the device.
Patent Information
- Application Number
- CN202410661028.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-30
- Filing Date
- 2024-05-27
- Publication Date
- 2024-12-03
AI Technical Summary
Existing semiconductor devices have poor mechanical properties during thermal or temperature cycles, thick metallization layers are harmful to electrical properties, making it difficult to achieve low Rds(on) and mechanical robustness at the same time.
By using a single additional lithography stage and selective etching technology, metallization layers of different thicknesses are configured, combined with interleaved layer structures, the number of photolithography masks is reduced, and the combination of thick metals in the active region and thin metals in the edge region is optimized, and metal yield is optimized. Strength and morphology.
Effectively reduce the number of photolithography masks of semiconductor devices in the active and edge regions, combine low Rds(on) and mechanical robustness, reduce passivation rupture and pattern shift, and improve the thermal mechanical stress stability of the device. .
Smart Images

Figure CN119069360A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method for manufacturing a semiconductor device. Background Art
[0002] Semiconductor devices, especially metal oxide semiconductor field effect transistor ("MOSFET") devices, are primarily designed as single-layer metallization devices in order to keep manufacturing costs low. The thickness of the metallization layer is guided by bonding force considerations for wire-bonded devices or by some aspects of diffusion resistance for clip-bonded devices. For certain specific applications, such as power MOSFETs, the metallization layer should be as thick as possible so that the device exhibits low on-resistance Rds(on) and uniform current spreading.
[0003] However, while thick metallization layers are electrically desirable, they are mechanically suboptimal because semiconductor devices are often manufactured with a passivation layer, such as Si3N4, SiO2, or a combination thereof, to protect the device from environmental contamination and surface damage, but which is susceptible to cracking during thermal or temperature cycling ("TMCL") reliability testing, particularly in edge regions. Thus, the desired electrical and mechanical properties of the metallization layer are in opposition to one another, as low Rds(on) values require thick metal layers, but mechanical robustness of the passivation layer requires a minimal amount of topography and favors thin metal layers.
[0004] Therefore, semiconductor devices are mainly manufactured using a single metal layer with a thickness optimized for low Rds(on) or for the mechanical robustness of the thin layer, and more rarely, using double layer metallization technology as a compromise to try and meet both requirements.
[0005] However, such dual-layer metallization typically requires at least two additional photolithography layers to accommodate the second metal layer and contact formation, as well as some form of chemical mechanical planarization (CMP) processing to keep the topography as flat as possible to mitigate cracking. The additional processing steps and costs associated with dual-layer metallization are not insignificant. Summary of the invention
[0006] Aspects of the invention are set out in the appended claims respectively with respect to various embodiments of a method of manufacturing a semiconductor device, and a semiconductor device manufactured by the method.
[0007] It is an object of the present invention to match the opposing requirements of low profile and thick metallization layers in a semiconductor device configured with metallization layers of varying thickness using a single additional lithography stage and stripping, and without the need for additional CMP processing.
[0008] Therefore, in a first aspect, a method for manufacturing a semiconductor device is provided, the method comprising the following steps: depositing a metal material layer onto a substrate, the layer having a top surface; masking at least an edge region of the layer with a first photolithography mask; removing metal material from the edge region according to the first mask, whereby a first portion of the edge region has a thickness between the substrate and the top surface of the layer; masking the first portion of the edge region with a second photolithography mask; and removing metal material from the first portion according to the second mask, whereby at least a second portion of the edge region is free of deposited metal material.
[0009] According to the solution, areas of semiconductor devices requiring low resistance, such as their active areas, can be configured with full metallization thickness, while areas of semiconductor devices sensitive to topography, such as their edge terminations, can be configured with reduced metallization thickness, wherein the reduction in metallization thickness reduction is achieved by only one lithographic mask and wet or dry removal, such as by etching. Thus, the disclosed solution advantageously reduces the number of lithographic masks required to deliver thick layers in the active areas of semiconductor devices, such as power MOSFETs, and thinner layers in adjacent edge areas, which effectively combines the characteristics of low Rds(on) and mechanical robustness.
[0010] Certain embodiments of the method may include the further step of configuring the second mask to define at least one feature in the first portion bounded by the adjacent second portion. Features so defined are advantageous for certain types of semiconductor devices, such as MOSFETs, that require at least one gate feature in the edge termination region.
[0011] In a variation of such an embodiment, the construction of the second mask may further include defining a second member in the first portion. The second member so defined is advantageous for certain types of semiconductor devices requiring at least one gate member and at least one source member in an edge termination region, wherein the gate member is located intermediate the source member and a layer adjacent to the first portion.
[0012] Certain embodiments of the method may comprise the further step of configuring the first mask to define the or each feature having a first width, wherein configuration of the second mask further comprises defining at least one of the or each feature having a second width greater than the first width.
[0013] In the case of any edge region configured with at least one gate member, the selective etching proposed in this embodiment can provide a gate busbar that maintains its full metal thickness, the same as the thickness of the active area of the device, and makes the thinner metallization layer have a greater impact on the gate resistance R of the device. G minimize the impact.
[0014] In the case of any edge region configured with gate and source components, the selective etching proposed in this embodiment can also provide both gate and source buss that help reduce passivation cracking and pattern shift because both maintain thicker metal at their center for optimal conductivity while maintaining thinner metal around the periphery of the structure where the thermomechanical stress is high. This results in a gate resistance R G The mechanical robustness of the device is significantly improved while minimizing the impact of the device on the mechanical stability.
[0015] In certain embodiments of the method, the depositing step may further include: interleaving at least one layer of a second material between the substrate and the top surface of the layer, the interleaved layer having a corresponding top surface; and the step of removing the metallic material from the edge region may further include: removing the metallic material whereby the first portion has a thickness between the substrate and the top surface of the interleaved layer. In such embodiments, the interleaved layer constitutes an intermediate layer endpoint detection barrier located intermediate the active region and the edge region.
[0016] In certain embodiments of the method, the deposition step may also include: interleaving a plurality of second material layers between the substrate and the top surface of the layer, each interleaved layer having a corresponding top surface; the method may also include the following steps: configuring a plurality of first masks in sequence for interleaving edge regions at each of the plurality of interleaved layers; and the step of removing the metal material from the edge region may also include: removing the metal material with each of the plurality of first masks.
[0017] In such an embodiment, each staggered layer constitutes an intermediate layer endpoint detection barrier located between the active area and the edge area, and a step is configured at each staggered layer, approximating a continuous shallow angle slope between the first and second portions and the active area to reduce the topography from the active area to the edge area while allowing thin metallization of one or more components in the edge area.
[0018] In a variation of such an embodiment, the step of configuring a plurality of first masks may further include configuring at least a last first mask in sequence to define at least two features in the first portion.
[0019] In order to maintain R G The method advantageously achieves pattern resist topography over the staggered gaps, defining the thickness of features in the edge region according to a predetermined thickness between the highest and lowest staggered layers. Depending on the number of staggered layers, the manufacturing sequence may thus include thinning down features, such as gate features or gate and source busbars, in the first, two or more etches to reduce thermomechanical stresses, but maintaining the metallization thickness in the subsequent one or two etches to maintain a low R G .
[0020] The second material for the interleaved layer may be selected from the first group including titanium, vanadium, chromium, zirconium, niobium, molybdenum, hafnium, tantalum, tungsten; or the second group including cobalt, nickel, copper, rhodium, palladium, silver, iridium, platinum, gold; or a group including alloys of the first group and / or the second group. In a variation of such an embodiment, the method may include an additional step of reacting the second material with oxygen (O2) or with nitrogen (N2).
[0021] In another aspect, there is provided a semiconductor device manufactured according to any of the techniques disclosed herein.
[0022] Further aspects of the invention are set out in the dependent claims. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The invention will be more clearly understood from the following description of embodiments of the invention given by way of example only with reference to the accompanying drawings, in which:
[0024] FIG. 1A shows details of a semiconductor device manufactured using a prior art method.
[0025] Figure 1B The equivalent detail of FIG. 1A is shown in a semiconductor device fabricated in accordance with the present invention.
[0026] Figure 2 The steps of the method of manufacturing a semiconductor device according to the first embodiment are described in detail.
[0027] Figure 3 A semiconductor device is shown at a stage during its manufacture by means of a second embodiment of the method, the semiconductor device comprising in a first portion a single component.
[0028] Figure 4 A semiconductor device is shown at a stage during its manufacture using a third embodiment of the method, the semiconductor device comprising two components in a first portion.
[0029] Figure 5 A semiconductor device comprising a member of varying width in a first portion is shown at stages during manufacture using a fourth embodiment of the method.
[0030] Figure 6 A semiconductor device comprising two members of varying width in a first portion is shown at a stage during manufacture using a fifth embodiment of the method.
[0031] Figure 7 Steps of another embodiment of a method of fabricating a semiconductor device including one or more staggered layers are described in detail.
[0032] Figure 8 Shown in accordance with Figure 7The method is directed to stages during the manufacture of a semiconductor device having a single staggered layer.
[0033] Fig. 9 Shown in accordance with Figure 7 The method is directed to stages during the manufacture of a semiconductor device having a plurality of staggered layers. DETAILED DESCRIPTION
[0034] The specific mode that the inventor envisions will now be described by way of example. In the following description and accompanying drawings, many specific details are set forth to provide a thorough understanding, wherein the same reference numerals represent the same features. It will be apparent to those skilled in the art that the present invention can be practiced without being limited to these specific details. In other cases, well-known methods and structures are not described in detail to avoid unnecessary fuzzy descriptions.
[0035] Referring first to FIG. 1A , a detail of a semiconductor device 10 manufactured using a prior art method is shown in a cross-sectional perspective view. The semiconductor device 10 includes a substrate 12 in cross section, which is optionally covered by an interlayer dielectric (“ILD”) layer 14, on the top surface of which a single metal layer 16, typically aluminum copper (AlCu), is deposited at a specific thickness of, for example, 3 μm. The deposition can be performed using any technique known to those skilled in the art, such as physical vapor deposition (“PVD”) techniques known in the specific field of semiconductor manufacturing, which are implemented as sputtering or evaporation. A photolithography process, such as described in the following paragraph, is optionally applied to the metal layer 16, and then a passivation layer 18, such as Si3N4, SiO2, or a combination thereof, is applied to the top surface of the device 10.
[0036] The detail shown in FIG. 1A is a lateral edge region of a prior art device 10 in which an active region 20 of the device, corresponding in this example to a source feature 22 of the semiconductor, is bounded by an edge region 30 or edge termination region including in this example a gate member 32 and a source member 34 .
[0037] In the prior art method, the gate member 32 and the source member 34 are typically formed by masking the metal layer 16 with a photolithographic mask that encodes the respective positions of the gate member 32 and the source member 34 relative to the active area 20, and then performing a wet or dry etching operation to remove the metal material from the layer 16 down to the ILD layer 14 according to the mask, whereby the gap portions 40 of the metal layer 16 are completely removed in the edge region 30, respectively, between the edge of the active area 20 and the gate member 32, between the gate member and the source member itself, and from the source member 34 to the edge of the substrate 12. Therefore, the active area 20 of the prior art semiconductor device 10 and the gate member 32 and the source member 34 all exhibit the same thickness, i.e., the vertical height relative to the surface of the substrate and the ILD layers 12, 14.
[0038] The inventors have empirically observed, based on comparisons of devices configured with 2 μm and 3 μm AlCu metal layers, respectively, and examined after 500 cycles on TMCL (dT=205°C), that during thermal cycling, thinner metal layers 16 produce fewer cracks in the passivation layer 18 at the extreme corners and metal edges of the device 10. The higher incidence of passivation cracks is attributed to a phenomenon known as the "yield strength" of the material, where the material begins to plastically deform, causing the material to separate or delaminate itself from its neighbors and thus sustain thermomechanical stresses alone in the device.
[0039] Therefore, the inventors have assumed the existence of a dimensionless quantity S, expressed as a function of the Young's modulus, yield strength, thermal expansion coefficient, Poisson's ratio and shear stress of the metal and silicon materials of the device considering temperatures THi and TLo, which is an indicator of the thermomechanical stability with respect to crack formation when the device is thermally cycled between temperatures THi and TLo, wherein yielding and ratcheting of AlCu will not occur under TMCL as long as S<2, and wherein the thermomechanical stresses caused by TMCL will be borne by all the materials constituting the body of the device, i.e. the passivation layer 18, the metal 16 and the device body 12. This stability criterion S<2 can be enhanced by increasing the yield strength value of the metal material, which in turn depends on the shear force acting on the material during TMCL, the width of the material and the thickness of the material.
[0040] Since the width of the material is determined by the dimensions of the device, the parameter that can be modified is the thickness of the metallization layer 16, and the inventors have observed that, through modeling based on a normalized thickness of 3 μm typically used in semiconductor devices, a reduction of AlCu from 3 μm to 2 μm should result in a 50% increase in the yield strength value of the metal material, while a reduction to 1 μm should result in a 300% increase.
[0041] Therefore, the inventors have designed a novel manufacturing method to reduce the thickness of the metallization layer 16 in the peripheral edge region of the semiconductor device, an exemplary embodiment of which is shown in FIG. Figure 1B 1A , and for ease of reference, mirrors the detail and perspective view of FIG. 1A , with like reference numerals corresponding to like features.
[0042] Figure 1B The semiconductor device 100 comprises in cross section the same substrate 12 optionally covered by an ILD layer 14, and in step 201, a single metal layer 16 is deposited on the top surface of the ILD layer 14, for example by sputtering a metal alloy with a thickness t1 of 3 μm again. A photolithography process according to the method of the present invention is then applied to the metal layer 16, and then a passivation layer 18, for example Si3N4, SiO2 or a combination thereof, is applied again to the top surface of the device 100, the photolithography process being as described below with reference to Figure 2 Describe and as in Figure 3 Targeted Figure 1B The semiconductor device 100 is shown.
[0043] Figure 1B The detail shown in is again a lateral edge region of the device 100 , wherein the active region 20 is still bounded by an edge region 30 or edge termination region which comprises a single gate member 32 in this example.
[0044] According to the method, after depositing the metal layer 16 at step 201 and before forming the gate member 32, the edge region 30 of the metal layer 16 is masked with a first photolithographic mask at step 202. A certain thickness of metal material is then removed from the edge region according to the first mask, for example using a dry or wet etching process, whereby a first portion 130 of the edge region has a remaining thickness t2 between the substrate 12 (optionally covered by the ILD layer 14) and the top surface of the metal layer 16, i.e., less than the thickness t1 of 3 μm of the metal layer 16 deposited at step 201, for example, 1 or 2 μm. The first portion 130 extends from the edge of the active area 20 immediately adjacent to the edge region 30, i.e., where the metal layer 16 transitions from its total thickness t1 to a reduced thickness t2 and effectively forms a step, towards the lateral edge of the device. The length of the first portion 130 may be in the range of 20 to 50 μm.
[0045] Then, in step 203, the topography of the edge region 30 in the first exemplary device 100 is formed by masking the first portion 130 of the edge region with a second photolithography mask, and then in step 204, removing material from the metal layer 16 in the first portion 130 according to the second mask over the remaining thickness t2, i.e., down to the top surface of the substrate layer 12 (or ILD layer 14, when present), thereby forming one or more second portions 40 of the metal layer 16 in the edge region 30. In this first example including the gate member 32, the second portion 40 is located between a certain distance from the step at the edge of the active region 20 and a side of the gate member 32 facing the active region 20, and from another second portion located on the opposite side of the gate member 34 to the edge of the substrate 12.
[0046] The active region 20 retains its original thickness t1 of the deposited metal layer 16 and is therefore optimized for reduced Rds(on). This region will typically be inside the active region of the device 100 and will extend substantially to the edge of the edge termination structure. The edge region 30, which in this example includes the gate member 32, is advantageously configured with a thinner thickness t2 and is therefore optimized for metal yield strength.
[0047] The steps between the active and edge regions 20, 30 are still susceptible to passivation cracking. However, in the case where a passivation crack occurs at the edge of the active region 20, the tensile energy will be released at the bottom of the step, where any resulting propagating crack will be dispersed over the edge region 30, which will absorb even the longest cracks observed in practice, given their length of 20 to 50 μm. Inter-region propagating cracks are considered benign because they propagate between portions of the same metal structure 16 without penetrating into underlying layers. Therefore, semiconductor devices fabricated using the present method are extremely robust to passivation cracking and uncontrolled release of thermomechanical energy.
[0048] The skilled person will appreciate that the method of the present invention can be implemented without configuring the edge region 30, without any other topography except the edge of the first portion 130 farthest from the active region 20, i.e., where the metal layer 16 transitions downward from its reduced thickness t2 to the top surface of the substrate layer 12 (or ILD layer 14, when present). Thus, after step 204, the single second portion 40 of the edge region 30 farthest from the active region 20 is free of deposited metal material.
[0049] The skilled person will also appreciate that the method of the present invention may be implemented with many different configurations of edge region 30, depending on the desired characteristics and functionality of the target semiconductor device. Figures 4 to 6 Presented and described as first referred to Figure 2 Variant embodiments of the method are described in which like reference numerals correspond to like features.
[0050] exist Figure 1B and Figure 3 In the embodiment of the device 100 shown, the edge region 30 is provided with a single gate member 32. In an embodiment of the method, as Figure 4 As shown, the edge region 30 of the semiconductor device 400 is configured with two components, namely a gate component 32 and a source component 34. As described above, a metal layer 16 is deposited at step 201, a first mask is applied at step 202, and material is removed from the metal layer 16 at step 203 according to the first mask.
[0051] In this embodiment, the second mask of step 204 is further configured to define a second source member 34 in addition to the first gate member 32, whereby, after the second mask is applied to the first portion 130 in step 204 and material is removed therefrom over the remaining thickness t2 according to the second mask in step 205, a second portion 40 is located between a distance from the step at the edge of the active region 20 and a side of the gate member 32 facing the active region 20, yet another second portion is located from the opposite side of the gate member 32 to the side of the source member 34 facing the active region 20, and yet another second portion is located from the opposite side of the source member 34 to the edge of the substrate 12. The edge region 30, which includes both the gate member 32 and the source member 34 in this example, is again advantageously configured with a thinner thickness t2, and is therefore optimized for metal yield strength.
[0052] In another embodiment of the method, Figure 5 As shown, the edge region 30 of the semiconductor device 500 is again configured with a gate member 32 and a source member 34. The gate member 32 is designed to enhance the gate resistance characteristic R of the semiconductor device 500 relative to the source member 34 with a reduced thickness t2. G and for this purpose preferably maintains a total thickness t1 of the metal layer 16 deposited in step 201 .
[0053] In this embodiment, the first mask of step 202 is additionally configured to define an upper portion of gate member 32 having a first width w1, and the second mask of step 204 is additionally configured to define both a second source member 34 and a lower portion of gate member 34 having a second width w2 greater than w1, noting that the second source member 34 can be configured to have the same larger width w2.
[0054] A metal layer 16 is deposited at step 201, a first mask is applied at step 202, and material is removed from the metal layer 16 according to the first mask at step 203, thus on either side of an upper portion of a gate member 32 having a first width w1. A second mask is applied at step 204, and material is removed according to the second mask at step 205 over a remaining thickness t2 in the first portion 130, thus wherein second portions 40 are located on one side of a lower portion of the gate member 32 having a larger width w2, on opposite sides of the lower portion of the gate member 32 and the center of one side of a source member 34, and on opposite sides of the source member 34, respectively. The edge region 30, including both the gate member 32 and the source member 34, is again advantageously configured with a thinner thickness t2, thus optimized for metal yield strength, whereas the gate member 32 has an overall thickness t1 and a wider lower portion that provides a thicker metal at its center to enhance conductivity.
[0055] In another embodiment of the method, Figure 6 As shown, the edge region 30 of the semiconductor device 600 is again configured with a gate member 32 and a source member 34, wherein in this embodiment, both members are designed as busbars, each having a total thickness t1 of the metal layer 16 as deposited in step 201. In this embodiment, the first mask of step 202 is additionally configured to define an upper portion of both the gate member 32 and the source member 34 having a first width w1, and the second mask of step 204 is additionally configured to define a lower portion of each of the gate member 32 and the source member 34 having a second width w2 greater than w1.
[0056] A metal layer 16 is deposited at step 201, a first mask is applied at step 202, and material is removed from the metal layer 16 according to the first mask at step 203, thus on either side of the respective upper portions of each of the gate member 32 and the source member 34. A second mask is applied at step 204, and material is removed at step 205 according to the second mask over the remaining thickness t2 in the first portion 130, whereby the second portion 40 is formed as previously described with reference to FIG. Figure 5 position.
[0057] The edge region 30, including both the gate member 32 and the source member 34, is still advantageously (at least partially in this embodiment) configured with a thinner thickness t2, thereby being optimized for metal yield strength, and each of the gate member and the source member 32 has both an overall thickness t1 and a wider lower portion that provides thicker metal at its center, thereby producing a device 600 that combines enhanced conductivity and improved mechanical robustness because the wider lower portion of each bus bar 32, 34 enables pattern shift and passivation cracking to be minimized in the edge region 30.
[0058] A further variation of the method contemplates interleaving one or more layers of a second material within the metal layer 16, wherein the or each such interleaved layer acts as an endpoint detection layer and allows for better control of the metallization thickness in the edge termination region 30. Although slightly more complex than the previous embodiment, the method advantageously maintains the advantages of the previous embodiment, namely maintaining the original thickness t1 of the metallization for optimal diffusion resistance in the power region 20 of the device, and providing a smaller thickness t2 of the metallization in its edge region 30 for enhanced mechanical robustness, and, depending on the number of interleaved layers, also allows the topography of the transition between the two regions to be smoothed, with the step at each intermediate layer approximating a continuous shallow angle slope.
[0059] Reference now Figure 7 A general embodiment of this variation is described, with specific examples in Figure 8 and 9 , where like reference numerals correspond to like features. In such a variation, the method step 201 of depositing includes a series of sub-steps, including at least a first sub-step 701 of depositing an initial sub-layer 810 of a metallic material onto the substrate 12 (or the ILD layer 14 on top thereof) with a thickness t3, which is a portion of the total thickness t1 of the metallization required for the active area 20, and in the example case of a single staggered layer, which may correspond to the reduced thickness t2 described in the previous embodiment. At least one instance of the next sub-step 702 involves depositing an staggered layer 820 or first staggered layer 820 of a second material (in the example, a titanium (Ti) or titanium nitride (TiN) alloy) with the minimum thickness allowed by the deposition technique used. At least one instance of the next sub-step 703 involves depositing a next sub-layer 810 of a metallic material onto the deposited layer of the second material 820 with a thickness t3 substantially the same as the initial sub-layer or another thickness.
[0060] Then, at sub-step 704, a question is asked as to whether the staggered metallization design of semiconductor device 800 requires another staggered layer 820 of a second material. In the affirmative, the method returns to sub-step 702, where another staggered layer 820 of a second material is deposited, and then at step 703, the next sub-layer 810 of a metallic material is deposited onto the second deposited layer of the second material 820, and so on. Alternatively or ultimately, the last deposited sub-layer 810 of metallic material achieves the total thickness t1 of the metallization required for active area 20, and the question of sub-step 704 is answered in the negative, where layer 16 includes one or more staggered layers 820 of a second material.
[0061] Control proceeds to masking with a first photolithography mask as previously described at step 202, followed by removal of material from metal layer 16 in edge region 30 according to the same method at step 203. The skilled person will appreciate that the masking and removal of steps 202, 2023 may include or exclude the interleaved layer 820 beneath the volume of removed metal material, depending on the desired properties of the semiconductor device. Figure 8 In the example shown, the staggered layer 820 is uniformly maintained within the first portion 130, including the top portions of the gate member 32 and the source member 34 of reduced thickness t2, so that the first portion has a thickness between the substrate and the top surface of the staggered layer. Fig. 9 In the example shown, the interleaved layer 820 remains on top of the gate member 32 and the source member 34 , but not on other aspects of the first portion 130 .
[0062] Furthermore, since the metal layer 16 may include more than one staggered layer 820, depending on the thickness t1 of the metal layer 16 before the first mask and the desired topography of the edge region 30 of the device 800, the method may require multiple first masks, the number of which depends on the number of staggered layers 820 within the metal layer 16 and the configuration of each first mask in its sequence, optionally defining staggering of the edge region 30 at each of the multiple staggered layers adjacent to the active 20, as Fig. 9 shown.
[0063] Those skilled in the art will appreciate that this variation is also fully compatible with the previously described embodiments, for example wherein one or more members 32, 34 may be confined within the first portion 130 of the edge region, with Figure 8 The reduced thickness t3 shown, or having a Fig. 9 The intermediate thickness shown is subject to at least the last first mask in the configuration sequence to define the or each feature in the first portion 130 before material is completely removed with the second mask in the second portion 40 down to the substrate 12 (or ILD layer 14).
[0064] Appropriately, the method then accordingly queries in an additional step 705 whether there are further interleaving layers 820 in the metal layer 16, for example interleaving layers 820 within the remaining thickness of the metal layer 16 which has not yet been removed, which require a further masking and removal cycle of steps 202, 203 and a corresponding and respective further first mask therefor. In the affirmative, the method returns to step 202, wherein a further first photolithography mask is applied to the edge region 30, and then in step 203 further material is removed from the metal layer 16 remaining in the first portion 130 according to the same method, and so on.
[0065] Optionally or eventually, the last sublayer of metal material 810 removed achieves the level of reduction in the metallization thickness t3 required for the edge region 20, and the question of substep 705 is answered in the negative, wherein the method proceeds to masking with a second photolithography mask as previously described at step 204, and then material is removed from the metal layer 16 in the first portion 130 according to the same method at step 205. The second portion 40 is again located a distance from the lowest step 830 of the edge region 30 and between the first component 32, intermediate any adjacent components, and from the opposite side of the component 34 farthest from the active region 20 to the edge of the substrate 12.
[0066] The presently disclosed manufacturing method provides a semiconductor device having a combination of reduced Rds(on) in its active region and improved mechanical robustness in its edge region, with predictable levels of passivation crack guidance and suppression. The presently disclosed manufacturing method can be used for all semiconductor devices made with known technology, such as silicon (Si) or compound semiconductors containing elements from groups III and V in the periodic table ("III-V compounds"), such as silicon carbide (SiC), gallium nitride (GaN), in which the passivation layer is prone to mechanical fatigue and failure due to cracks that extend to the underlying layer through thermal mismatch between the different materials constituting the device. Such semiconductor devices include various power devices (MOSFETs, bipolar junction transistors (BJT), insulated gate bipolar transistors (IGBT), etc.) and integrated power devices (driver and MOSFET modules (DrMOS), temperature and overload protection field effect transistors (TopFET), etc.).
[0067] In the specification, the terms "comprise, comprise, comprised and comprising" or any variation thereof and the terms "include, include, included and including" or any variation thereof are considered to be fully interchangeable, and they should all be given the broadest possible interpretation, and vice versa. The present invention is not limited to the embodiments described above, but may be varied in structure and detail.
Claims
1. A method for manufacturing a semiconductor device, comprising the following steps: depositing a layer of metallic material onto the substrate, the layer having a top surface; masking at least an edge region of the layer with a first photolithographic mask; removing metallic material from the edge region according to the first mask, whereby a first portion of the edge region has a thickness between the substrate and a top surface of the layer; masking the first portion of the edge region with a second photolithography mask; as well as Metal material is removed from the first portion according to the second mask, whereby at least a second portion of the edge region is free of deposited metal material.
2. The method according to claim 1, further comprising the steps of: The second mask is configured to define at least one feature in the first portion, wherein the or each feature is bounded by an adjacent second portion. 3 . The method of claim 2 , wherein the at least one member defined in the first portion is a gate member.
4. The method of claim 3, wherein at least a second component defined in the first portion is a source component.
5. The method according to any one of claims 2 to 4, further comprising the following steps: configuring the first mask to define the or each feature having a first width; The step of configuring the second mask further comprises defining at least one of the or each feature having a second width greater than the first width.
6. The method according to any one of claims 1 to 5, wherein: The depositing step further includes interleaving at least one layer of a second material between the substrate and the layer top surface, the interleaved layers having respective top surfaces; and The step of removing metallic material from the edge region further includes removing metallic material whereby the first portion has a thickness between the substrate and the top surface of the interleaved layer.
7. The method according to claims 1 to 5, wherein: The depositing step further includes interleaving multiple layers of a second material between the substrate and the layer top surface, each interleaved layer having a corresponding top surface; The method further comprises the steps of: sequentially configuring a plurality of first masks for staggering the edge region at each of the plurality of staggered layers; and The step of removing metal material from the edge region further includes removing metal material using each of the plurality of first masks.
8. The method according to claim 7, wherein: The step of configuring a plurality of first masks further includes configuring at least a last first mask in the sequence to define the at least two features in the first portion.
9. The method according to claim 7 or 8, wherein the second material is selected from The first group consisting of titanium, vanadium, chromium, zirconium, niobium, molybdenum, hafnium, tantalum and tungsten; or The second group includes cobalt, nickel, copper, rhodium, palladium, silver, iridium, platinum and gold.
10. The method according to claim 9, wherein the second material is selected from the group consisting of alloys of the first group and / or the second group.
11. Method according to claim 9 or 10, comprising a further step of reacting the second material with oxygen (O2) or with nitrogen (N2).
12. A semiconductor device manufactured using the method according to any one of claims 1 to 11. 13 . The semiconductor device according to claim 12 , which is made of silicon (Si) or a III-V compound semiconductor.
14. The semiconductor device according to claim 12 or 13, selected from the group consisting of MOSFET, bipolar junction transistor (BJT), insulated gate bipolar transistor (IGBT), driver and MOSFET module (DrMOS), temperature and overload protection field effect transistor (TopFET).
